Throughout the world, 12 million people acquire syphilis every year. About one-third reside in sub-Saharan Africa, where the proportion of individuals with syphilis is 2-3 times higher in those who are HIV-infected compared to those who are not. In the US, compared to all patients with primary and secondary syphilis, rates of primary and secondary syphilis are 80 times higher in patients who are HIV-infected. Development of early neurosyphilis is more common in HIV-infected individuals than in those who are not HIV infected. Early neurosyphilis causes substantial morbidity, including symptomatic meningitis, hearing loss, vertigo, and visual loss. Neurosyphilis begins with asymptomatic meningitis. Untreated, asymptomatic neurosyphilis can lead to symptomatic neurosyphilis. Identification and treatment of the ~25% of HIV-infected patients with asymptomatic neurosyphilis would prevent them from progressing to symptomatic disease. However, there is currently no sensitive and specific test to identify these individuals. One approach to this problem would be to perform a lumbar puncture on all HIV-infected patients with syphilis. However this approach is impractical in many settings and potentially wastes resources. In this application, we build upon our previous work in a cohort of 556 HIV-infected subjects with syphilis to develop and validate algorithms to 1) determine which HIV- infected patients with syphilis should undergo lumbar puncture and 2) interpret CSF abnormalities. These algorithms will be constructed using a novel statistical method called classification and regression tree (CART) analysis. Further, we will determine the impact of concomitant antiretroviral therapy, and of particular antiretroviral agents within a regimen, on response to neurosyphilis therapy. Our study is unique. It is the only study of neurosyphilis conducted in the HIV era. Neurosyphilis carries significant morbidity in HIV-infected patients. The studies proposed in this application will yield the necessary tools for clinicians to optimally evaluate, treat and manage HIV-infected patients with syphilis and neurosyphilis. Syphilis, a disease that many thought was more or less controlled, is enjoying a major come back in the developed world and remains endemic in the developing world. Many of those infected also have HIV, and they are at increased risk of a complication called neurosyphilis, which can cause blindness, deafness and stroke within weeks after syphilis infection. Our proposed research builds on our previous work and will lead to ways to better diagnose and treat neurosyphilis.